ROLE OF FILLERS IN THE FATIGUE BEHVIOUR OF A SHORT GLASS FIBRE REINFORCED POLYAMIDE Andrea Bernasconi Politecnico di Milano, Dipartimento di Meccanica Via La Masa 1, I-20156 Milano, Italy andrea.bernasconi@polimi.it Corrado Armanni Radici Novacips Chignolo d'isola (Bg) Italy corrado.armanni@radicigroup.com ABSTRACT Fatigue tests on different types of reinforced polyamides were conducted with the aim of understanding their fatigue behaviour and the role of the different types of fillers and reinforcements on the fatigue damage mechanism. Fillers and reinforcements consisted of short glass fibres, mineral filler, glass beads and mixtures of these reinforcements. The fatigue properties were analyzed with reference to tensile strength values and Young s moduli. Polyamide reinforced with only short glass fibres displayed the highest fatigue resistance in spite of having the lowest tensile stiffness, whereas the use of mineral fillers allowed for increasing the stiffness but reduced considerably the tensile and the fatigue strength. The analysis of fracture surfaces allowed for identifying the features typical of fatigue damage nucleation sites, consisting of matrix filaments having failed in a ductile manner. The position of these filaments, in close proximities of debonded filler particles, supports the assumption that mineral fillers and glass beads, although contributing to the increase of the composite s stiffness, act as stress raisers, thus accelerating crack formation. 1. INTRODUCTION Short glass fibre reinforced polyamides are often employed for load bearing applications in substitution of metals. The fatigue behaviour of short glass fibre polyamides has already been thoroughly studied [1], particularly with the aim of understanding the role of short glass fibres (orientation [2], length [3] and notches [4]). The mechanism of fatigue crack nucleation in SGFR polyamide has been thoroughly described by Horst [1,5], who proposed a model assuming that crack nucleation starts at fibre ends, where high shear stresses concentrate, causing the appearance of voids at fibre ends; then cracks propagate along the fibres by failure of the fibre-matrix interface, until fibres are completely debonded, thus leaving filaments of highly deformed matrix. These filaments keep bridging the crack faces, thus making crack propagation cohesive, until the crack reaches a critical length. Finally, failure of the remaining portion of the specimen s section takes place in a brittle manner. The phases of this mechanism of fatigue crack formation are depicted in Figure 1. Short glass fibres are not the only possibility for reinforcing thermoplastics like polyamides. Other fillers, like mineral fillers (talc, kaolin), or glass beads are often added to polyamides, in order to tailor their properties to specific applications [6-8]. The fatigue behaviour of mineral filled thermoplastics has been reported in the literature only in the case of polypropylene [9]. In this work, results of fatigue tests on a SGFR polyamide 6 filled with mineral filler, glass beads and a mixture of these fillers with short glass fibres are presented and
compared with tensile properties. The possible mechanisms of fatigue damage nucleation are discussed on the basis of fracture surface observations. Figure 1: Fatigue crack nucleation mechanism of SGFR polyamide (illustration from Ref. [1]): 1, debonding starts at fibre ends; 2, voids surface propagate along fibrematrix interface; 3, matrix filaments form; 4, when filaments fail, a crack is formed.. 2. EXPERIMENTS Tensile and fatigue tests were performed on a polyamide 6 with the following reinforcements and fillers: 10% by weight short glass fibres (PA6-GF10), 10% by weight short glass fibres and 20% mineral filler (PA6-(GF10+M20)), 10% by weight short glass fibres and 20% glass beads (PA6-(GF10+GB20)) and 23% mineral filler (PA6-M23). The 23% weight fraction of PA6-M23 corresponds to the fraction of filler contained in the matrix of PA6-(GF10+M20). Talc was used as mineral filler. Standard specimens were injection moulded into type 1A shape, as defined by the ISO 527-2 standard [10] and shown in Figure 2. Tensile tests were conducted at 1 mm/min crosshead speed. Fatigue tests were run with a stress ratio R = σ min /σ max of 0.1, at a frequency of 2 Hz. Failure criterion was specimen separation. Tests were conducted at an ambient temperature of 23 C. At the beginning of tests, the materials were in the dry as moulded state. Figure 2: Size and dimensions of ISO 537-2 type 1A specimen.
3. RESULTS AND DISCUSSION 3.1 Tensile testing The relevant tensile properties of the materials are reported in Table 1. By adding mineral filler to the composite, the stiffness is considerably increased, whereas the tensile strength is reduced. Adding glass beads to glass fibre reinforced polyamide increases the Young s modulus and leaves tensile strength practically unaltered. Table 1: Tensile properties of glass fibre reinforced and mineral or glass beads filled polyamide 6 Material Elastic modulus (MPa) Stress at break (MPa) Strain at break (%) PA6-GF10 4540 100 3,0 PA6-(GF10+M20) 6630 85 2,8 PA6-(GF10+GB20) 5780 101 3,8 PA6-M23 5190 74 2,6 Figure 3: Specific (solid bars) and absolute (dashed bars) values of the tensile strength of glass fibre reinforced and mineral or glass beads filled polyamide 6. Figure 4: Specific (solid bars) and absolute (dashed bars) values of the tensile modulus of glass fibre reinforced and mineral or glass beads filled polyamide 6.
Specific values, i.e. divided by density, of the tensile strength and Young s modulus are reported in the graphs of Figure 3 and Figure 4, respectively. It clearly appears that in the case of PA6-(GF10+GB20), when compared with PA6-GF10, the use of glass beads diminishes the specific tensile strength, whereas the absolute values remain unchanged. Fillers like mineral particles and glass beads tend to reduce specific strength, whereas both specific and absolute values of tensile modulus are increased. 3.2 Fatigue testing The fatigue results are reported in Figure 5, where the log(maximum applied fatigue stress), or log σ max, is plotted vs. log (cycles to failure) or log N f. The stress-life curves (also known as S N curves or Woehler curves), interpolating the experimental results are also shown. Based on the above mentioned experimental results, the following equation was used to describe the relationship between maximum stress and cycles to failure: b σ max = σ f N (1) The values of fatigue strength exponent b and fatigue strength coefficients σ f are listed in Table 2. Figure 5: S-N curves of glass fibre reinforced and mineral or glass beads filled polyamide 6. Table 2: Fatigue properties of glass fibre reinforced and mineral or glass beads filled polyamide 6 Material Fatigue strength coefficient (MPa) Fatigue strength exponent (-) PA6-GF10 142-0.0569 PA6-(GF10+GB20) 127.7-0.0586 PA6-(GF10+M20) 102.5-0.0438 PA6-M23 94.1-0.0429
Although giving a relevant contribution to the composite s stiffness, it appears that the mineral filler acts as a local stress raiser at the micro-scale, thus determining a decrease in both tensile and fatigue strength. Glass beads have a similar effect as mineral filler at low stress levels, corresponding to approximately 10 6 cycles to failure, whereas mineral particles affect the fatigue strength of glass fibre reinforced polyamide more severely at shorter lives. By comparing the slopes of the S-N curves, it appears that the curves of PA6-M23 and PA6-(GF+M20) are parallel to each other, and different than that of PA6-GF10, suggesting that fatigue behaviour is mainly controlled by the presence of the mineral filler. Moreover, the addition of glass beads to short glass fibre reinforced polyamide causes a reduction of strength under fatigue loading, which is not observed in tensile tests. The slope value of the S-N curve of PA6-(GF10+GB20) equal to that of PA6- GF10 indicates that a damage mechanism involving glass beads could have been activated, similar but less severe than that of mineral filler, presumably because of the higher stress concentration at the borders of the mineral particles. 3.3 Fractography By observing the fracture surfaces of fatigued specimens, it appears that the fatigue mechanism in polyamide reinforced by fibres only is similar to that described by Horst [1], characterized by cohesive cracking starting at fibre ends, reported in the Introduction. Fracture surfaces compatible with this mechanism were observed in the specimens of PA6-GF10. The surface of a specimen observed at SEM is shown in Figure 6, where the micro-ductile area is highlighted. Figure 6: Typical fracture surface of a fatigued specimen of PA6 GF10 (88 MPa maximum stress); the red line shows the perimeter of the micro-ductile area. This area corresponds to the nucleation and cohesive propagation of the crack. As shown in Figure 7, in this area fibres are completely debonded, indicating that failure of the fibre-matrix interface took place, and numerous filaments of highly drawn matrix are visible, which presumably had bridged the cack face during crack propagation.
Figure 7: Detail of the micro-ductile area of fatigued PA6 GF10. The remaining portion of the fracture surface displays a completely different aspect, usually indicated as micro-brittle, which is very similar to that of specimens which failed in tensile tests. A detail of this area is shown in Figure 8, where the following differences can be observed: some fibres are covered by a thin layer of matrix, indicating that in this case the matrix, and not the interface, failed, and the matrix shows little signs of deformation. Figure 8: Detail of the micro-brittle area of fatigued PA6 GF10. By observing the fracture surfaces of the mineral filled and glass fibre reinforced polyamide (Figure 9), micro-ductile (Figure 9 a) and micro-brittle (Figure 9 b) areas can still be distinguished, thus allowing for identifying the crack nucleation zone. However, it appears that mineral filled materials displayed a different cracking behaviour than the same material reinforced with fibres only; crack nucleation took
place at the smaller scale of mineral particles, which appear debonded from the matrix, whereas glass fibres are still adhering to the matrix. (a) (b) Figure 9: Detail of the micro-ductile (a) and micro brittle (b) area of a specimen of PA6-(GF10+M20) fatigued at 73 MPa. A similar appearance of the crack surface was found in the samples of mineral filled polyamide 6 (Figure 10).Thus it may be inferred that mineral particles, presumably because of their lamellar shape inducing a higher local stress concentration than fibre ends, constitute the weakest link for crack nucleation in the mineral filled materials of this work, thus explaining the severe strength reduction. Figure 10: Detail of the micro-ductile area of a specimen of PA6-M23 fatigued at 68 MPa. A similar mechanism can be inferred from the observation of the micro-ductile area of specimens of PA6-(GF10+GB20), where glass beads are likely to have been the crack initiation sites. Crack initiation in proximity of glass bead presumably took place before the fibre-matrix interface failed. This assumption is supported by the observation of
several fibres still encased in the matrix in Figure 11. The higher fatigue strength of PA6-(GF10+GB20), as compared with PA6-(GF10-M20), can be explained in terms of lower local stress concentration induced by glass beads (whose aspect ratio is equal to one) than that of mineral particles, characterized by a much higher aspect ratio. Figure 11: Detail of the micro-ductile area of the fracture surface of a specimen of PA6- (GF10+GB20) fatigued at 68 MPa. 4. CONCLUSIONS Tensile and fatigue tests were conducted on different types of reinforced polyamide 6 with the aim of understanding the role of different types of fillers and reinforcements on the fatigue damage mechanism. Fillers and reinforcements consisted of glass fibres (PA6-GF10), mineral filler (PA6-M23), mineral filler and glass fibres (PA6- (GF10+M20) and glass fibres with glass beads (PA6-(GF10+GB20)). Polyamide reinforced with only short glass fibres displayed the highest fatigue resistance in spite of having the lowest tensile stiffness, whereas the use of mineral filler allowed for increasing the stiffness but reduced considerably the tensile and the fatigue strength. Glass beads left tensile strength unaltered, but reduced the fatigue lives. The analysis of fracture surfaces allowed for identifying the features typical of fatigue damage nucleation sites in reinforced polyamides, consisting of matrix filaments having failed in a ductile manner. However these features appeared at different scales, i.e. at the scale of the fibres in PA6-GF10, and at the smaller scale of filler particles in the case of the other materials. The position of these filaments in close proximities of filler particles support the assumption that mineral fillers and glass beads, although contributing to the increase of the composite s stiffness, act as stress raisers, thus accelerating crack formation. Particles of small size reduce the time of crack nucleation, and the lamellar shape of mineral particles constitute a more severe weakening of the matrix, as compared with glass beads. ACKNOWLEDGEMENTS The authors wish to thank Guido Sessa for help in conducting the experiments.
REFERENCES 1- Horst JJ, Spoormaker JL. Fatigue fracture mechanism and fractography of short-glassfibre-reinforced polyamide 6. Journal of Material Science 32 (1997) 3641-3651 2- Bernasconi A, Davoli P, Basile A, Filippi A. Effect of fibre orientation on the fatigue behaviour of a short glass fibre reinforced polyamide-6. Int J Fatigue 29 (2007): 199-208 3- Bernasconi A, Davoli P, Rossin D, Armanni C. Analysis of the effect of mechanical recycling upon tensile strength of a short glass fibre reinforced polyamide 6,6. Engineering Fracture Mechanics, 74 (2007): 627-641 4- Sonsino CM, Moosbrugger E. Fatigue design of highly loaded short-glass-fibre reinforced polyamide parts in engine compartments. Int J Fatigue (2007), doi: 10.1016/j.ijfatigue.2007.08.017 5- Horst JJ, Spoormaker JL. Mechanism of fatigue in short glass fiber reinforced polyamide 6. Polymer Engineering and Science 36 (1996): 2718-2726 6- Huang L, Yuan Q, Jiang W, An L, Jiang S, Li RKY. Mechanical and thermal properties of glass bead filled nylon 6. Journal of Applied Polymer Science 94 (2004): 1885-1890 7- Unal H. Morphology and mechanical properties of composites based on polyamide 6 and mineral additives. Materials and Design 25 (2004): 483-487 8- Unal H, Findik F, Mimaroglu A. Mechanical behavior of nylon composites containing talc and kaolin. Journal of Applied Polymer Science 88 (2003): 1694-1697 9- Zhou Y, Mallick PK. Fatigue performance of an injection molded talc-filled polypropylene. Polymer Engineering and Science (2005): 511-516 10- ISO 527-2. Plastics Determination of tensile properties Part 2: Test conditions for moulding and extrusion plastics. International Organization For Standards (1993)